No abstract

This paper describes REVIEW, a new retinal vessel reference dataset. This dataset includes 16 images with 193 vessel segments, demonstrating a variety of pathologies and vessel types. The vessel edges are marked by three observers using a special drawing tool. The paper also describes the algorithm used to process these segments to produce vessel profiles, against which vessel width measurement algorithms can be assessed. Recommendations are given for use of the dataset in performance assessment. REVIEW can be downloaded from http://ReviewDB.lincoln.ac.uk.

This is the first case series describing central scotoma associated with silicone oil in situ. Electrophysiology indicated macular dysfunction in most cases. We suggest that early removal of oil in cases with good visual potential should be considered to avoid this sight-threatening complication.

and for the BONSAI (Brain and Optic Nerve Study with Artificial Intelligence) Study Group Objective: To compare the diagnostic performance of an artificial intelligence deep learning system with that of expert neuro-ophthalmologists in classifying optic disc appearance. Methods: The deep learning system was previously trained and validated on 14,341 ocular fundus photographs from 19 international centers. The performance of the system was evaluated on 800 new fundus photographs (400 normal optic discs, 201 papilledema [disc edema from elevated intracranial pressure], 199 other optic disc abnormalities) and compared with that of 2 expert neuro-ophthalmologists who independently reviewed the same randomly presented images without clinical information. Area under the receiver operating characteristic curve, accuracy, sensitivity, and specificity were calculated. Results: The system correctly classified 678 of 800 (84.7%) photographs, compared with 675 of 800 (84.4%) for Expert 1 and 641 of 800 (80.1%) for Expert 2. The system yielded areas under the receiver operating characteristic curve of 0.97 (95% confidence interval [CI] = 0.96-0.98), 0.96 (95% CI = 0.94-0.97), and 0.89 (95% CI = 0.87-0.92) for the detection of normal discs, papilledema, and other disc abnormalities, respectively. The accuracy, sensitivity, and specificity of the system's classification of optic discs were similar to or better than the 2 experts. Intergrader agreement at the eye level was 0.71 (95% CI = 0.67-0.76) between Expert 1 and Expert 2, 0.72 (95% CI = 0.68-0.76) between the system and Expert 1, and 0.65 (95% CI = 0.61-0.70) between the system and Expert 2. Interpretation: The performance of this deep learning system at classifying optic disc abnormalities was at least as good as 2 expert neuro-ophthalmologists. Future prospective studies are needed to validate this system as a diagnostic aid in relevant clinical settings.

PurposeIntravitreal anti-vascular endothelial growth factor (VEGF) agents are effective in the treatment of central involving diabetic macular oedema (DMO). Vitreoretinal interface abnormalities (VRIA) are common in patients with DMO, and the effect of these on the response to anti-VEGF treatment is unclear. Furthermore the effect of anti-VEGF agents on the VRIA itself is uncertain.MethodProspective study of consecutive patients treated with ranibizumab (RZB) for DMO as part of routine clinical care in one eye unit over a 1-year period. Visual acuity (Va), central retinal thickness (CRT) and injection frequency data was recorded on an electronic database. Treatment was initiated with four monthly RZB injections and then a monthly PRN regime. All patients underwent high-density spectral-domain optical coherence tomography (SDOCT) at baseline and 12 months. The SDOCTs were graded by two observers masked to the outcome.ResultsOne hundred and four eyes (77 patients) were included in the analysis. The mean age was 62 years, and 62% were male. The mean presenting vision was 62 letters and CRT 472 μm. Eighty eyes retained stable Va, and 17 had an improvement in Va. At baseline, 39 eyes had associated focal vitreomacular adhesion (VMA) and by 12 months this reduced to 30 (p = 0.04), with 12 releasing VMA and three developing it. Patients with VMA had significantly better final Va than those without VMA. Improvement in CRT was greatest in those where VMA released during the study. Forty-five eyes had some degree of foveal involving epiretinal membrane (ERM) at baseline, and 28 were considered to have clinically significant ERM. There was no clinically relevant change in ERM during the study. Patients with significant ERM at baseline had a lower final vision. Multivariate analysis showed that ERM and more severe retinopathy at baseline were predictive of less visual improvement (p < 0.01). Shorter intraretinal cyst length, ERM and the absence of VMA at baseline were predictive of a worsened anatomical response (p < 0.001).ConclusionVRIA are related to outcome in patients treated with RZB. ERM was associated with a worsened visual and anatomic response, and VMA with an improved anatomical response particularly when spontaneous VMA release occurred during treatment. The presence and severity of ERM was not affected by RZB treatment.Electronic supplementary materialThe online version of this article (doi:10.1007/s00417-016-3562-0) contains supplementary material, which is available to authorized users.